Autoimmune pancreatitis (AIP) is increasingly recognised in adults (1). Clinical features include obstructive jaundice, radiological signs of a diffuse pancreatic enlargement and irregular narrowing of the pancreatic duct, increased serum levels of immunoglobulin G4 (IgG4), variable presence of autoantibodies, and clinical response to steroids (1). An international consensus statement was published suggesting 2 distinct variants: type 1 AIP, seen in approximately 85% of adult patients and characterized by elevated serum IgG4 concentrations and association with IgG4-related disease at extrapancreatic sites (eg, sialadenitis, retroperitoneal fibrosis), and type 2 AIP, in which the serological markers are usually absent with no systemic manifestations except for possible association with inflammatory bowel disease (2). Histologically, type 1 AIP shows massive infiltration by IgG4+ plasma cells, whereas the type 2 AIP is characterized by neutrophilic infiltration with granulocytic epithelial lesions (GELs) and few, if any, IgG4+ plasma cells (1–4). We describe our experience in diagnosing AIP in 2 children and discuss how to improve diagnostic approach to this condition.
An 11-year-old boy presented with epigastric pain, jaundice, and weight loss. Blood tests showed bilirubin 225 μmol/L (normal range 3–20), γ-glutamyl transpeptidase 120 IU/L (1–55), alkaline phosphatase 597 IU/L (120–488), aspartate aminotransferase 78 IU/L (10–50), and amylase 33 IU/L (<100). Serum IgG4 was normal at 37.6 mg/dL (<135 mg/dL). Autoantibodies were negative. No mutations for hereditary pancreatitis were detected. On imaging, the pancreas appeared diffusely enlarged with loss of the cobblestone architecture of the pancreatic surface (Fig. 1A). No discrete mass was identified. Endoscopic retrograde pancreatocholangiography showed a tight stricture in the lower bile duct and almost occluded pancreatic duct. To rule out malignancy, he underwent a computed tomography (CT)–guided biopsy, which demonstrated a severe sclerosing inflammation. In addition to the lymphoplasmacytic infiltration, many neutrophils were noted, some infiltrating the lining epithelium of pancreatic ducts/ductules, consistent with GELs (Fig. 1B). IgG4+ plasma cells were scarce on immunostaining. Type 2 AIP was provisionally diagnosed and the patient was treated with prednisolone (1 mg · day−1 · kg−1), which dramatically reduced the pancreatic enlargement and the biochemical abnormalities. The steroid dose was weaned gradually and stopped after 12 months. The patient remains off steroids in remission after >2 years of follow-up.
A 14-year-old girl presented with a 2-week history of jaundice and pale stools. Blood tests showed bilirubin 82 μmol/L, γ-glutamyl transpeptidase 452 IU/L, alkaline phosphatase 672 IU/L, aspartate aminotransferase 261 IU/L, and amylase 39 IU/L. Apart from myeloperoxidase-anti-neutrophilic cytoplasmic antibody at 8 U/mL (normal range <5), serum autoantibodies were negative. IgG4 was slightly raised at 190 mg/dL. CT revealed a mass-like enlargement of the pancreatic head with a tight stricture of the lower bile duct and calibre irregularity of the pancreatic duct. Endoscopic papillotomy and insertion of the biliary stent improved her jaundice. A CT-guided pancreatic biopsy demonstrated lymphoplasmacytic infiltration, periductal and lobular neutrophilic infiltration with GELs, and no IgG4+ plasma cells on immunostaining, the features consistent with type 2 AIP (Fig. 1C, D). Steroids were not given because the biochemical changes improved and repeated endoscopic retrograde pancreatocholangiography for the stent removal showed resolution of the pancreatic abnormalities. One year later, the girl developed melaena, and following colonoscopy, was diagnosed as having ulcerative colitis. Initial treatment with prednisolone, mesalazine, and azathioprine failed to control her symptoms and the patient eventually required adalimumab to induce remission. No further episodes of jaundice or pancreatitis have been recorded during 7 years.
We describe 2 paediatric cases of AIP, in which the diagnosis was only reached following histological confirmation of the specific inflammation in the pancreas. Radiological features in our patients were not specific. They have certainly not included a peripancreatic “capsule-like” rim (1), suggested to be specific for AIP in adults. Radiological findings in AIP of childhood appear to be less dramatic than in adults. The differential diagnosis in children includes hereditary pancreatitis, inflammatory pseudotumour, and atypical presentation of cystic fibrosis.
Recent recommendations from the International Consensus Diagnostic Criteria (5) suggest that only above twice the normal levels of serum IgG4 should be considered highly suggestive of type 1 AIP, based on the previous observation that serum IgG4 levels can be nonspecifically elevated up to 280 mg/dL in various inflammatory and neoplastic conditions of the pancreas (6). In the absence of serological markers, the diagnosis of type 2 AIP is more challenging, requiring a histological confirmation of GELs (5). In contrast to adult patients, in which suspected pancreatic enlargement could indicate malignancy and often requires prompt tissue diagnosis, in children, pancreatic biopsies are performed less frequently, which could contribute to the difference in described cases. In our previous pilot study, in the absence of histological evidence, we diagnosed only 1 AIP among 30 children with chronic pancreatitis during 20 years (7).
Only 16 children with AIP have been reported in the literature (Table 1) (4,8). Similar to adult patients, their key clinical finding was a diffuse steroid-responsive pancreatic enlargement; however, some other features appear to be distinct. For example, type 1 AIP with IgG4 elevation >280 mg/dL and/or extrapancreatic sclerosing lesions has never been identified in the paediatric population. If International Consensus Diagnostic Criteria are applied to the reported paediatric cases, patients 1 to 8 would be diagnosed as having “definitive type 2 AIP” based on histological evidence of GEL, the same as our 2 cases. Patient 9 would belong to “probable type 2 AIP” based on history of Crohn disease (8). Patients 10 to 14 would be categorized as “AIP, not otherwise specified.” In those children, serum IgG4 was completely normal. Patients 15 and 16 would be categorised as “probable type 1 AIP,” because their diagnosis is supported by either slightly higher serum IgG4 level (224 mg/dL) or histological features, but without definite evidence for type 1 AIP. A caveat is that neutrophilic infiltration and GELs have not been considered important until recently and not consistently sought for in the literature.
In conclusion, the predominant type of AIP in children appears to be type 2, given that more than a half of reported paediatric cases had GELs, whereas “definitive type 1 AIP” has never been described in children. For paediatric patients with suspected AIP, diagnostic process should be targeted more towards type 2, in which the tissue diagnosis is recommended.
1. Park DH, Kim MH, Chari ST. Recent advances in autoimmune pancreatitis. Gut
2. Chari ST, Kloeppel G, Zhang L, et al. Histopathologic and clinical subtypes of autoimmune pancreatitis: the Honolulu consensus document. Pancreatology
3. Notohara K, Burgart LJ, Yadav D, et al. Idiopathic chronic pancreatitis with periductal lymphoplasmacytic infiltration: clinicopathologic features of 35 cases. Am J Surg Pathol
4. Zamboni G, Lüttges J, Capelli P, et al. Histopathological features of diagnostic and clinical relevance in autoimmune pancreatitis: a study on 53 resection specimens and 9 biopsy specimens. Virchows Arch
5. Shimosegawa T, Chari ST, Frulloni L, et al. International consensus diagnostic criteria for autoimmune pancreatitis: guidelines of the International Association of Pancreatology. Pancreas
6. Ghazale A, Chari ST, Smyrk TC, et al. Value of serum IgG4 in the diagnosis of autoimmune pancreatitis and in distinguishing it from pancreatic cancer. Am J Gastroenterol
7. Grammatikopoulos T, Davies E, Davenport D, et al. Does Autoimmune Pancreatitis exist in childhood? A twenty year single centre retrospective review (abstract). 43rd Annual Meeting of the European Association for the Study of the Liver Disease, Milan, Italy. J Hepatol
© 2014 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,
8. Fujii LL, Chari ST, El-Youssef M, et al. Pediatric pancreatic EUS-guided trucut biopsy for evaluation of autoimmune pancreatitis. Gastrointest Endosc